AimIn recent years, with the concept of damage control resuscitation, hemostasis and preoperative fluid restriction have been carried out, but there is controversy regarding the effectiveness of fluid restriction.MethodsFrom April 2007 to March 2013, 101 trauma patients presented with hemorrhagic shock (systolic blood pressure ≤90 mmHg) at the prehospital or emergency department and were admitted to Hyogo Emergency Medical Center (Hyogo, Japan). They underwent emergency hemostasis by surgery and transcatheter arterial embolization. We compared two groups in a historical cohort study, the aggressive fluid resuscitation (AR) group, which included 59 cases treated in the period April 2007–March 2010, and the fluid restriction (FR) group, which included 42 cases treated in the period April 2010–March 2013.ResultsThere was no difference between both groups in patient background (heart rate, 110 b.p.m.; systolic blood pressure, 70 mmHg). The Injury Severity Score was 34 (AR) versus 38 (FR) (not significant). Preoperative infusion volume of crystalloid significantly decreased, from 2310 mL (AR) to 1025 mL (FR) (P ≤ 0.01). There was no difference in mortality (36% [AR] versus 41% [FR]). Ventilator days significantly decreased, from 8.5 days (AR) to 5.5 days (FR) (P = 0.02).ConclusionsPreoperative fluid restriction for trauma patients with hemorrhagic shock did not improve mortality, but it decreased ventilator days by reducing the perioperative plus water balance and it might contribute to perioperative intensive care.